Photovoltaic power generation system fault monitoring method and system

Abstract

The invention relates to the technical field of fault monitoring, and discloses a photovoltaic power generation system fault monitoring method and system, and the method comprises the steps: carrying out the multi-angle shooting of a to-be-fault-monitored photovoltaic power generation system, obtaining a plurality of multi-angle images, splicing the multi-angle images, and obtaining a to-be-fault-analyzed panoramic image; performing gray level conversion on the panoramic image to be subjected to fault analysis to obtain a panoramic gray level image to be subjected to fault analysis, and determining a preset standard gray level image; analyzing the panoramic grayscale image to be subjected to fault analysis and a preset standard grayscale image, and determining an image difference coefficient; and judging whether the photovoltaic power generation system to be subjected to fault monitoring has a fault risk according to the image difference coefficient, and giving an alarm prompt when the photovoltaic power generation system to be subjected to fault monitoring has the fault risk, thereby realizing real-time monitoring and intelligent diagnosis of the global health condition of the photovoltaic power generation system, improving the accuracy and efficiency of fault monitoring, and improving the reliability of fault monitoring. And a powerful guarantee is provided for stable operation of a photovoltaic power generation system.

Description

Technical Field

[0001] The present invention relates to the technical field of fault monitoring, and in particular to a method and system for monitoring faults in a photovoltaic power generation system. Background Art

[0002] With the continuous growth of global energy demand and the increasing awareness of environmental protection, photovoltaic power generation, as a clean and renewable energy utilization method, is receiving more and more attention and application. Photovoltaic power generation systems provide a green and sustainable energy solution for mankind by converting solar energy into electrical energy. However, the stable operation of photovoltaic power generation systems is crucial to ensure the stability of energy supply, which requires real-time and effective fault monitoring of photovoltaic modules and related equipment.

[0003] Existing photovoltaic power generation system fault monitoring methods mainly focus on the detection of single components. Although these methods can detect faults to a certain extent, they have shortcomings such as limited monitoring range, slow response speed, and inability to achieve global monitoring. Especially in large-scale photovoltaic power stations, due to the large number of equipment and their wide distribution, traditional monitoring methods are difficult to meet the needs of fast and accurate fault location. Summary of the invention

[0004] The embodiments of the present invention provide a photovoltaic power generation system fault monitoring method and system, which realize real-time monitoring and intelligent diagnosis of the global health status of the photovoltaic power generation system, realize early warning and precise positioning of potential faults, improve the accuracy and efficiency of fault monitoring, and provide a strong guarantee for the stable operation of the photovoltaic power generation system.

[0005] In order to achieve the above object, the present invention provides a photovoltaic power generation system fault monitoring method, comprising:

[0006] Determine a photovoltaic power generation system to be monitored for faults, photograph the photovoltaic power generation system to be monitored for faults from multiple angles to obtain multiple multi-angle images, and stitch the multi-angle images to obtain a panoramic image to be analyzed for faults;

[0007] Performing grayscale conversion on the panoramic image to be analyzed for faults to obtain the panoramic grayscale image to be analyzed for faults, and determining a preset standard grayscale image of the photovoltaic power generation system to be monitored for faults;

[0008] Analyzing the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image, and determining an image difference coefficient between the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image based on the analysis result;

[0009] It is determined whether the photovoltaic power generation system to be monitored for failure has a failure risk according to the image difference coefficient, and an alarm is issued when the photovoltaic power generation system to be monitored for failure has a failure risk.

[0010] Furthermore, when the photovoltaic power generation system to be monitored for fault is photographed at multiple angles to obtain multiple multi-angle images, and the multi-angle images are stitched to obtain a panoramic image to be analyzed for fault, the method includes:

[0011] Preprocessing all multi-angle images, wherein the preprocessing includes denoising, contrast enhancement and geometric correction;

[0012] Extract key feature points from each multi-angle image based on the scale-invariant feature transformation detection algorithm;

[0013] According to the KD-tree feature matching algorithm, the key feature points in each multi-angle image are matched by nearest neighbor search to find the corresponding feature point pairs, and the wrong feature point pairs are deleted according to the RANSAC algorithm;

[0014] Determine the homography matrix between the multi-angle images through the geometric relationship between the feature point pairs, and align the multi-angle images based on the homography matrix;

[0015] The registered multi-angle images are fused and spliced ​​to obtain the panoramic image to be analyzed for the fault.

[0016] Furthermore, when analyzing the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image, and determining the image difference coefficient between the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image based on the analysis result, it includes:

[0017] Extracting pixel values ​​of pixel points in the panoramic grayscale image to be analyzed for faults, and clustering all pixel values ​​to determine pixel value clusters;

[0018] Determine the cluster center of the pixel value cluster, and determine the pixel distance from each pixel value to the cluster center;

[0019] Determine an average pixel distance and a boundary pixel distance of all pixel distances, and calculate an overall attribute factor corresponding to each pixel point based on the average pixel distance and the boundary pixel distance;

[0020] The overall attribute factor corresponding to each pixel is calculated according to the following formula:

[0021]

[0022] Among them, e is the overall attribute factor corresponding to the i-th pixel, r iis the pixel distance corresponding to the i-th pixel point, f is the average pixel distance, g is the boundary pixel distance, a1 is the first calculation coefficient, a2 is the second calculation coefficient, a3 is the third calculation coefficient, and a1+a2+a3=1, a1>a2>a3;

[0023] Analyze all the overall attribute factors, and determine multiple sub-overall attribute factor sequences of the panoramic grayscale image to be analyzed for faults based on the analysis results;

[0024] An image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image is calculated based on a plurality of sub-overall attribute factor sequences.

[0025] Furthermore, when all the overall attribute factors are analyzed and multiple sub-overall attribute factor sequences of the panoramic grayscale image to be analyzed for fault are determined based on the analysis results, it includes:

[0026] All the overall attribute factors are sorted according to the numerical value relationship, and the maximum overall attribute factor and the adjacent overall attribute factors form the first initial sub-overall attribute factor sequence, and the third overall attribute factor is used as the overall attribute factor to be updated;

[0027] Calculate the first sum of two overall attribute factors in the first initial sub-overall attribute factor sequence, and calculate the first difference between the overall attribute factor to be updated and the first sum, and determine whether the first difference is greater than or equal to a preset difference. If so, add the overall attribute factor to be updated to the first initial sub-overall attribute factor sequence, and update the first initial sub-overall attribute factor sequence;

[0028] If not, take the first initial sub-overall attribute factor sequence as a sub-overall attribute factor sequence, obtain the fourth overall attribute factor, combine the third overall attribute factor and the fourth overall attribute factor to form the second initial sub-overall attribute factor sequence, and determine a new overall attribute factor to be updated;

[0029] Calculate the second sum of the two overall attribute factors in the second initial sub-overall attribute factor sequence, and calculate the second difference between the new overall attribute factor to be updated and the second sum, and determine whether the second difference is greater than or equal to the preset difference. If so, add the new overall attribute factor to be updated to the second initial sub-overall attribute factor sequence, and update the second initial sub-overall attribute factor sequence;

[0030] If not, take the second initial sub-integral attribute factor sequence as a sub-integral attribute factor sequence;

[0031] In this way, updating and iteration are performed to obtain multiple sub-sequences of overall attribute factors of the panoramic grayscale image to be analyzed for faults according to all overall attribute factors.

[0032] Furthermore, when calculating the image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image based on a plurality of sub-whole attribute factor sequences, it includes:

[0033] Determine the maximum sub-integral attribute factor and the minimum sub-integral attribute factor in each sub-integral attribute factor sequence;

[0034] Calculate the range value between the maximum sub-whole attribute factor and the minimum sub-whole attribute factor;

[0035] Output the pixel score of each pixel point corresponding to the sub-whole attribute factor based on the pre-selected trained score model;

[0036] The sequence factor of each sub-whole attribute factor sequence of the panoramic grayscale image to be analyzed for fault is calculated according to the following formula:

[0037]

[0038] Among them, d is the sequence factor, h is the extreme value, hmin is the minimum sub-integral attribute factor, hmax is the maximum sub-integral attribute factor, n is the number of sub-integral attribute factors in the sub-integral attribute factor sequence, k is i is the pixel score of the pixel point corresponding to the ith sub-integral attribute factor in the sub-integral attribute factor sequence;

[0039] Determine a plurality of sequence factors corresponding to the preset standard grayscale image;

[0040] Perform curve fitting on each sub-whole attribute factor of the panoramic grayscale image to be analyzed for fault and a plurality of sequence factors corresponding to the preset standard grayscale image, respectively, to obtain a panoramic grayscale image fitting curve and a preset standard grayscale image fitting curve;

[0041] Determine whether there is an intersection between the panoramic grayscale image fitting curve and the preset standard grayscale image fitting curve, and if so, mark the intersection as qualified;

[0042] Perform unqualified matching on the remaining sequence factors to generate multiple unqualified matching pairs;

[0043] Calculating the factor sum and factor difference of two sequence factors in the unqualified matching pair;

[0044] An image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image is calculated according to the factor sum and the factor difference.

[0045] Further, when calculating the image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image according to the factor sum value and the factor difference value, it includes:

[0046] The image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image is calculated according to the following formula:

[0047]

[0048] Among them, w is the image difference coefficient, m is the number of unqualified matching pairs, q1 j is the factor and value of the jth unqualified matching pair, q2 j is the factor difference of the jth unqualified matching pair, ((q1j-q2j) 2 )min is the minimum value among all (q1j-q2j), ((q1j-q2j) 2 )max is the maximum value among all (q1j-q2j), y 2 For all (q1 j -q2 j ) 2 The variance of .

[0049] Further, when judging whether the photovoltaic power generation system to be monitored for fault has a fault risk according to the image difference coefficient, it includes:

[0050] Obtaining a preset image difference coefficient, and judging whether the photovoltaic power generation system to be monitored for fault has a fault risk according to a relationship between the image difference coefficient and the preset image difference coefficient;

[0051] When the image difference coefficient is less than the preset image difference coefficient, it is determined that there is no failure risk in the photovoltaic power generation system to be monitored for failure;

[0052] When the image difference coefficient is greater than or equal to the preset image difference coefficient, it is determined that the photovoltaic power generation system to be monitored for fault has a fault risk.

[0053] In order to achieve the above object, the present invention also provides a photovoltaic power generation system fault monitoring system, comprising:

[0054] The first module is used to determine the photovoltaic power generation system to be monitored for faults, photograph the photovoltaic power generation system to be monitored for faults from multiple angles to obtain multiple multi-angle images, and stitch the multi-angle images to obtain a panoramic image to be analyzed for faults;

[0055] The second module is used to perform grayscale conversion on the panoramic image to be analyzed for faults, obtain the panoramic grayscale image to be analyzed for faults, and determine a preset standard grayscale image of the photovoltaic power generation system to be monitored for faults;

[0056] The third module is used to analyze the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image, and determine the image difference coefficient between the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image based on the analysis result;

[0057] The fourth module is used to determine whether the photovoltaic power generation system to be monitored for failure has a failure risk according to the image difference coefficient, and to issue an alarm reminder when the photovoltaic power generation system to be monitored for failure has a failure risk.

[0058] Compared with the prior art, the present invention has the following beneficial effects:

[0059] The present invention discloses a photovoltaic power generation system fault monitoring method and system, which comprises the following steps: performing multi-angle shooting on a photovoltaic power generation system to be monitored for fault to obtain a plurality of multi-angle images, and splicing the multi-angle images to obtain a panoramic image to be analyzed for fault; performing grayscale conversion on the panoramic image to be analyzed for fault to obtain a panoramic grayscale image to be analyzed for fault, and determining a preset standard grayscale image; analyzing the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image to determine an image difference coefficient; judging whether the photovoltaic power generation system to be monitored for fault has a fault risk according to the image difference coefficient, and issuing an alarm reminder when the photovoltaic power generation system to be monitored for fault has a fault risk, thereby realizing real-time monitoring and intelligent diagnosis of the global health status of the photovoltaic power generation system, improving the accuracy and efficiency of fault monitoring, and providing a strong guarantee for the stable operation of the photovoltaic power generation system. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Various other advantages and benefits will become apparent to those of ordinary skill in the art by reading the detailed description of the preferred embodiments below. The accompanying drawings are only for the purpose of illustrating the preferred embodiments and are not to be considered as limiting the present invention. Moreover, the same reference symbols are used throughout the accompanying drawings to represent the same components. In the accompanying drawings:

[0061] Figure 1 A schematic diagram of a photovoltaic power generation system fault monitoring method according to an embodiment of the present invention is shown;

[0062] Figure 2 A schematic diagram of the structure of a photovoltaic power generation system fault monitoring system in an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0063] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.

[0064] In the description of the present application, it should be understood that the terms "center", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc., indicating orientations or positional relationships, are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present application.

[0065] The terms "first" and "second" are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of the indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of the features. In the description of this application, unless otherwise specified, "plurality" means two or more.

[0066] In the description of this application, it should be noted that, unless otherwise clearly specified and limited, the terms "installed", "connected", and "connected" should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in this application can be understood according to specific circumstances.

[0067] The following is a description of preferred embodiments of the present invention with reference to the accompanying drawings.

[0068] like Figure 1 As shown, an embodiment of the present invention discloses a photovoltaic power generation system fault monitoring method, comprising:

[0069] S110: determining a photovoltaic power generation system to be monitored for a fault, photographing the photovoltaic power generation system to be monitored for a fault from multiple angles to obtain multiple multi-angle images, and stitching the multi-angle images to obtain a panoramic image to be analyzed for a fault;

[0070] In some embodiments of the present application, when the photovoltaic power generation system to be monitored for fault is photographed at multiple angles to obtain multiple multi-angle images, and the multi-angle images are stitched to obtain a panoramic image to be analyzed for fault, the method includes:

[0071] Preprocessing all multi-angle images, wherein the preprocessing includes denoising, contrast enhancement and geometric correction;

[0072] Extract key feature points from each multi-angle image based on the scale-invariant feature transformation detection algorithm;

[0073] According to the KD-tree feature matching algorithm, the key feature points in each multi-angle image are matched by nearest neighbor search to find the corresponding feature point pairs, and the wrong feature point pairs are deleted according to the RANSAC algorithm;

[0074] Determine the homography matrix between the multi-angle images through the geometric relationship between the feature point pairs, and align the multi-angle images based on the homography matrix;

[0075] The registered multi-angle images are fused and spliced ​​to obtain the panoramic image to be analyzed for the fault.

[0076] In this embodiment, preprocessing:

[0077] Denoising: Reduce noise in images and improve image quality.

[0078] Enhance contrast: Adjust the contrast of the image to make the image features more obvious and facilitate subsequent processing.

[0079] Geometric correction: Correct geometric distortion in images, such as lens distortion.

[0080] In this embodiment, feature point extraction: key feature points are extracted from each multi-angle image using a scale-invariant feature transform (SI FT) algorithm. SI FT is a method used for image recognition, which achieves this goal by finding and comparing feature points in different images.

[0081] In this embodiment, feature point matching: the KD-tree feature matching algorithm is used to perform nearest neighbor search matching to find the corresponding feature point pairs. KD-tree is a data structure that can efficiently query the nearest neighbor points in multidimensional space. The RANSAC algorithm is used to delete the wrong feature point pairs. RANSAC is an iterative method for estimating the parameters of a mathematical model, which can effectively exclude outliers (i.e., points that do not conform to the model).

[0082] In this embodiment, homography matrix determination and registration: The homography matrix between the multi-angle images is determined by the geometric relationship between the feature point pairs. The homography matrix describes the linear transformation relationship between two two-dimensional planes. The multi-angle images are registered based on the homography matrix, that is, they are aligned to the same coordinate system.

[0083] In this embodiment, fusion and stitching: the registered multi-angle images are fused and stitched to obtain the panoramic image to be analyzed for the fault. This step usually involves merging the information of multiple images into a single, continuous image to provide a more comprehensive perspective.

[0084] The beneficial effect of the above technical solution is: the present invention performs multi-angle shooting on the photovoltaic power generation system to be monitored for faults, obtains multiple multi-angle images, and splices the multi-angle images to obtain a panoramic image to be analyzed for faults, which can ensure the fault analysis of the entire photovoltaic power generation system and avoid the singleness of the fault analysis process.

[0085] S120: performing grayscale conversion on the panoramic image to be analyzed for faults to obtain the panoramic grayscale image to be analyzed for faults, and determining a preset standard grayscale image of the photovoltaic power generation system to be monitored for faults;

[0086] In this embodiment, the grayscale conversion technology is complex and mature and is not specifically limited here.

[0087] S130: Analyze the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image, and determine an image difference coefficient between the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image based on the analysis result;

[0088] In some embodiments of the present application, when analyzing the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image, and determining the image difference coefficient between the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image based on the analysis result, the method includes:

[0089] Extracting pixel values ​​of pixel points in the panoramic grayscale image to be analyzed for faults, and clustering all pixel values ​​to determine pixel value clusters;

[0090] Determine the cluster center of the pixel value cluster, and determine the pixel distance from each pixel value to the cluster center;

[0091] Determine an average pixel distance and a boundary pixel distance of all pixel distances, and calculate an overall attribute factor corresponding to each pixel point based on the average pixel distance and the boundary pixel distance;

[0092] The overall attribute factor corresponding to each pixel is calculated according to the following formula:

[0093]

[0094] Among them, e is the overall attribute factor corresponding to the i-th pixel, r i is the pixel distance corresponding to the i-th pixel point, f is the average pixel distance, g is the boundary pixel distance, a1 is the first calculation coefficient, a2 is the second calculation coefficient, a3 is the third calculation coefficient, and a1+a2+a3=1, a1>a2>a3;

[0095] Analyze all the overall attribute factors, and determine multiple sub-overall attribute factor sequences of the panoramic grayscale image to be analyzed for faults based on the analysis results;

[0096] An image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image is calculated based on a plurality of sub-overall attribute factor sequences.

[0097] In this embodiment, the demarcation pixel distance is to determine the median, that is, the median of all pixel distances.

[0098] The beneficial effect of the above technical solution is: the present invention determines the average pixel distance and the boundary pixel distance of all pixel distances, and calculates the overall attribute factor corresponding to each pixel point based on the average pixel distance and the boundary pixel distance, which can lay a foundation for determining the sub-overall attribute factor sequence.

[0099] In some embodiments of the present application, when analyzing all the overall attribute factors and determining multiple sub-overall attribute factor sequences of the panoramic grayscale image to be analyzed for faults based on the analysis results, it includes:

[0100] All the overall attribute factors are sorted according to the numerical value relationship, and the maximum overall attribute factor and the adjacent overall attribute factors form the first initial sub-overall attribute factor sequence, and the third overall attribute factor is used as the overall attribute factor to be updated;

[0101] Calculate the first sum of two overall attribute factors in the first initial sub-overall attribute factor sequence, and calculate the first difference between the overall attribute factor to be updated and the first sum, and determine whether the first difference is greater than or equal to a preset difference. If so, add the overall attribute factor to be updated to the first initial sub-overall attribute factor sequence, and update the first initial sub-overall attribute factor sequence;

[0102] If not, take the first initial sub-overall attribute factor sequence as a sub-overall attribute factor sequence, obtain the fourth overall attribute factor, combine the third overall attribute factor and the fourth overall attribute factor to form the second initial sub-overall attribute factor sequence, and determine a new overall attribute factor to be updated;

[0103] Calculate the second sum of the two overall attribute factors in the second initial sub-overall attribute factor sequence, and calculate the second difference between the new overall attribute factor to be updated and the second sum, and determine whether the second difference is greater than or equal to the preset difference. If so, add the new overall attribute factor to be updated to the second initial sub-overall attribute factor sequence, and update the second initial sub-overall attribute factor sequence;

[0104] If not, take the second initial sub-integral attribute factor sequence as a sub-integral attribute factor sequence;

[0105] In this way, updating and iteration are performed to obtain multiple sub-sequences of overall attribute factors of the panoramic grayscale image to be analyzed for faults according to all overall attribute factors.

[0106] The beneficial effect of the above technical solution is that the present invention can accurately obtain multiple sub-entire attribute factor sequences, thereby providing reliable data support for the calculation of image difference coefficients.

[0107] In some embodiments of the present application, when calculating the image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image based on a plurality of sub-whole attribute factor sequences, it includes:

[0108] Determine the maximum sub-integral attribute factor and the minimum sub-integral attribute factor in each sub-integral attribute factor sequence;

[0109] Calculate the range value between the maximum sub-whole attribute factor and the minimum sub-whole attribute factor;

[0110] Output the pixel score of each pixel point corresponding to the sub-whole attribute factor based on the pre-selected trained score model;

[0111] The sequence factor of each sub-whole attribute factor sequence of the panoramic grayscale image to be analyzed for fault is calculated according to the following formula:

[0112]

[0113] Among them, d is the sequence factor, h is the extreme value, hmin is the minimum sub-integral attribute factor, hmax is the maximum sub-integral attribute factor, n is the number of sub-integral attribute factors in the sub-integral attribute factor sequence, k is i is the pixel score of the pixel point corresponding to the ith sub-integral attribute factor in the sub-integral attribute factor sequence;

[0114] Determine a plurality of sequence factors corresponding to the preset standard grayscale image;

[0115] Perform curve fitting on each sub-whole attribute factor of the panoramic grayscale image to be analyzed for fault and a plurality of sequence factors corresponding to the preset standard grayscale image, respectively, to obtain a panoramic grayscale image fitting curve and a preset standard grayscale image fitting curve;

[0116] Determine whether there is an intersection between the panoramic grayscale image fitting curve and the preset standard grayscale image fitting curve, and if so, mark the intersection as qualified;

[0117] Perform unqualified matching on the remaining sequence factors to generate multiple unqualified matching pairs;

[0118] Calculating the factor sum and factor difference of two sequence factors in the unqualified matching pair;

[0119] An image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image is calculated according to the factor sum and the factor difference.

[0120] In this embodiment, the score model is pre-trained, and is trained based on different data samples and the scores matched by the data samples.

[0121] In this embodiment, the calculation method of the multiple sequence factors corresponding to the preset standard grayscale image is consistent with the calculation method of the sequence factors of each sub-overall attribute factor sequence of the panoramic grayscale image to be analyzed for faults, which will not be repeated here.

[0122] In this embodiment, each unqualified matching pair includes two sequence factors.

[0123] The beneficial effect of the above technical solution is that the present invention calculates the image difference coefficient between the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image according to the factor sum value and the factor difference, which can ensure the accuracy of fault analysis of the photovoltaic power generation system, avoid errors, and ensure intelligence and accuracy.

[0124] In some embodiments of the present application, when calculating the image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image according to the factor sum value and the factor difference value, it includes:

[0125] The image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image is calculated according to the following formula:

[0126]

[0127] Among them, w is the image difference coefficient, m is the number of unqualified matching pairs, q1 j is the factor and value of the jth unqualified matching pair, q2 j is the factor difference of the jth unqualified matching pair, ((q1 j -q2 j ) 2 ) min For all (q1 j -q2 j ), ((q1 j -q2 j ) 2 ) max For all (q1 j -q2 j ), y 2 For all (q1 j -q2 j ) 2 The variance of .

[0128] S140: judging whether the photovoltaic power generation system to be monitored for failure has a failure risk according to the image difference coefficient, and issuing an alarm reminder when the photovoltaic power generation system to be monitored for failure has a failure risk.

[0129] In some embodiments of the present application, when judging whether the photovoltaic power generation system to be monitored for failure has a failure risk according to the image difference coefficient, it includes:

[0130] Obtaining a preset image difference coefficient, and judging whether the photovoltaic power generation system to be monitored for fault has a fault risk according to a relationship between the image difference coefficient and the preset image difference coefficient;

[0131] When the image difference coefficient is less than the preset image difference coefficient, it is determined that there is no failure risk in the photovoltaic power generation system to be monitored for failure;

[0132] When the image difference coefficient is greater than or equal to the preset image difference coefficient, it is determined that the photovoltaic power generation system to be monitored for fault has a fault risk.

[0133] The beneficial effects of the above technical solution are: the present invention realizes real-time monitoring and intelligent diagnosis of the global health status of the photovoltaic power generation system, improves the accuracy and efficiency of fault monitoring, and provides a strong guarantee for the stable operation of the photovoltaic power generation system.

[0134] In order to further explain the technical idea of ​​the present invention, the technical solution of the present invention is now described in combination with specific application scenarios.

[0135] Correspondingly, such as Figure 2 As shown, the present application also provides a photovoltaic power generation system fault monitoring system, comprising:

[0136] The first module is used to determine the photovoltaic power generation system to be monitored for faults, photograph the photovoltaic power generation system to be monitored for faults from multiple angles to obtain multiple multi-angle images, and stitch the multi-angle images to obtain a panoramic image to be analyzed for faults;

[0137] The second module is used to perform grayscale conversion on the panoramic image to be analyzed for faults, obtain the panoramic grayscale image to be analyzed for faults, and determine a preset standard grayscale image of the photovoltaic power generation system to be monitored for faults;

[0138] The third module is used to analyze the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image, and determine the image difference coefficient between the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image based on the analysis result;

[0139] The fourth module is used to determine whether the photovoltaic power generation system to be monitored for failure has a failure risk according to the image difference coefficient, and to issue an alarm reminder when the photovoltaic power generation system to be monitored for failure has a failure risk.

[0140] In the description of the above embodiments, specific features, structures, materials or characteristics may be combined in a suitable manner in any one or more embodiments or examples.

[0141] Although the present invention has been described above with reference to the embodiments, various modifications may be made thereto and parts thereof may be replaced with equivalents without departing from the scope of the present invention. In particular, as long as there is no structural conflict, the various features in the embodiments disclosed by the present invention may be used in combination with each other in any manner, and the fact that these combinations are not fully described in this specification is only for the sake of omitting space and saving resources.

[0142] Those skilled in the art can understand that the above are only preferred embodiments of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art can still modify the technical solutions recorded in the aforementioned embodiments or replace some of the technical features therein by equivalents. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A photovoltaic power generation system fault monitoring method, characterized in that: include: Determine a photovoltaic power generation system to be monitored for faults, photograph the photovoltaic power generation system to be monitored for faults from multiple angles to obtain multiple multi-angle images, and stitch the multi-angle images to obtain a panoramic image to be analyzed for faults; Performing grayscale conversion on the panoramic image to be analyzed for faults to obtain the panoramic grayscale image to be analyzed for faults, and determining a preset standard grayscale image of the photovoltaic power generation system to be monitored for faults; Analyzing the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image, and determining an image difference coefficient between the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image based on the analysis result; It is determined whether the photovoltaic power generation system to be monitored for failure has a failure risk according to the image difference coefficient, and an alarm is issued when the photovoltaic power generation system to be monitored for failure has a failure risk.

2. The photovoltaic power generation system fault monitoring method according to claim 1, characterized in that: When the photovoltaic power generation system to be monitored for fault is photographed at multiple angles to obtain multiple multi-angle images, and the multi-angle images are spliced ​​to obtain a panoramic image to be analyzed for fault, the method includes: Preprocessing all multi-angle images, wherein the preprocessing includes denoising, contrast enhancement and geometric correction; Extract key feature points from each multi-angle image based on the scale-invariant feature transformation detection algorithm; According to the KD-tree feature matching algorithm, the key feature points in each multi-angle image are matched by nearest neighbor search to find the corresponding feature point pairs, and the wrong feature point pairs are deleted according to the RANSAC algorithm; Determine the homography matrix between the multi-angle images through the geometric relationship between the feature point pairs, and align the multi-angle images based on the homography matrix; The registered multi-angle images are fused and spliced ​​to obtain the panoramic image to be analyzed for the fault.

3. The photovoltaic power generation system fault monitoring method according to claim 1, characterized in that: When analyzing the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image, and determining the image difference coefficient between the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image based on the analysis result, the method includes: Extracting pixel values ​​of pixel points in the panoramic grayscale image to be analyzed for faults, and clustering all pixel values ​​to determine pixel value clusters; Determine the cluster center of the pixel value cluster, and determine the pixel distance from each pixel value to the cluster center; Determine an average pixel distance and a boundary pixel distance of all pixel distances, and calculate an overall attribute factor corresponding to each pixel point based on the average pixel distance and the boundary pixel distance; The overall attribute factor corresponding to each pixel is calculated according to the following formula: Among them, e is the overall attribute factor corresponding to the i-th pixel, r i is the pixel distance corresponding to the i-th pixel point, f is the average pixel distance, g is the boundary pixel distance, a1 is the first calculation coefficient, a2 is the second calculation coefficient, a3 is the third calculation coefficient, and a1+a2+a3=1, a1>a2>a3; Analyze all the overall attribute factors, and determine multiple sub-overall attribute factor sequences of the panoramic grayscale image to be analyzed for faults based on the analysis results; An image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image is calculated based on a plurality of sub-overall attribute factor sequences.

4. The photovoltaic power generation system fault monitoring method according to claim 3, characterized in that: When all the overall attribute factors are analyzed and multiple sub-overall attribute factor sequences of the panoramic grayscale image to be analyzed for fault are determined based on the analysis results, it includes: All the overall attribute factors are sorted according to the numerical value relationship, and the maximum overall attribute factor and the adjacent overall attribute factors form the first initial sub-overall attribute factor sequence, and the third overall attribute factor is used as the overall attribute factor to be updated; Calculate the first sum of two overall attribute factors in the first initial sub-overall attribute factor sequence, and calculate the first difference between the overall attribute factor to be updated and the first sum, and determine whether the first difference is greater than or equal to a preset difference. If so, add the overall attribute factor to be updated to the first initial sub-overall attribute factor sequence, and update the first initial sub-overall attribute factor sequence; If not, take the first initial sub-overall attribute factor sequence as a sub-overall attribute factor sequence, obtain the fourth overall attribute factor, combine the third overall attribute factor and the fourth overall attribute factor to form the second initial sub-overall attribute factor sequence, and determine a new overall attribute factor to be updated; Calculate the second sum of the two overall attribute factors in the second initial sub-overall attribute factor sequence, and calculate the second difference between the new overall attribute factor to be updated and the second sum, and determine whether the second difference is greater than or equal to the preset difference. If so, add the new overall attribute factor to be updated to the second initial sub-overall attribute factor sequence, and update the second initial sub-overall attribute factor sequence; If not, take the second initial sub-integral attribute factor sequence as a sub-integral attribute factor sequence; In this way, updating and iteration are performed to obtain multiple sub-sequences of overall attribute factors of the panoramic grayscale image to be analyzed for faults according to all overall attribute factors.

5. The photovoltaic power generation system fault monitoring method according to claim 4, characterized in that: When calculating the image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image based on a plurality of sub-whole attribute factor sequences, it includes: Determine the maximum sub-integral attribute factor and the minimum sub-integral attribute factor in each sub-integral attribute factor sequence; Calculate the range value between the maximum sub-whole attribute factor and the minimum sub-whole attribute factor; Output the pixel score of each pixel point corresponding to the sub-whole attribute factor based on the pre-selected trained score model; The sequence factor of each sub-whole attribute factor sequence of the panoramic grayscale image to be analyzed for fault is calculated according to the following formula: Among them, d is the sequence factor, h is the extreme value, hmin is the minimum sub-integral attribute factor, hmax is the maximum sub-integral attribute factor, n is the number of sub-integral attribute factors in the sub-integral attribute factor sequence, k is i is the pixel score of the pixel point corresponding to the ith sub-integral attribute factor in the sub-integral attribute factor sequence; Determine a plurality of sequence factors corresponding to the preset standard grayscale image; Perform curve fitting on each sub-whole attribute factor of the panoramic grayscale image to be analyzed for fault and a plurality of sequence factors corresponding to the preset standard grayscale image, respectively, to obtain a panoramic grayscale image fitting curve and a preset standard grayscale image fitting curve; Determine whether there is an intersection between the panoramic grayscale image fitting curve and the preset standard grayscale image fitting curve, and if so, mark the intersection as qualified; Perform unqualified matching on the remaining sequence factors to generate multiple unqualified matching pairs; Calculating the factor sum and factor difference of two sequence factors in the unqualified matching pair; An image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image is calculated according to the factor sum and the factor difference.

6. The photovoltaic power generation system fault monitoring method according to claim 5, characterized in that: When calculating the image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image according to the factor sum value and the factor difference value, it includes: The image difference coefficient between the panoramic grayscale image to be analyzed for faults and the preset standard grayscale image is calculated according to the following formula: Among them, w is the image difference coefficient, m is the number of unqualified matching pairs, q1 j is the factor and value of the jth unqualified matching pair, q2 j is the factor difference of the jth unqualified matching pair, ((q1 j -q2 j ) 2 ) min For all (q1 j -q2 j ), ((q1 j -q2 j ) 2 ) max For all (q1 j -q2 j ), y 2 For all (q1 j -q2 j ) 2 The variance of .

7. The photovoltaic power generation system fault monitoring method according to claim 1, characterized in that: When judging whether the photovoltaic power generation system to be monitored for fault has a fault risk according to the image difference coefficient, it includes: Obtaining a preset image difference coefficient, and judging whether the photovoltaic power generation system to be monitored for fault has a fault risk according to a relationship between the image difference coefficient and the preset image difference coefficient; When the image difference coefficient is less than the preset image difference coefficient, it is determined that there is no failure risk in the photovoltaic power generation system to be monitored for failure; When the image difference coefficient is greater than or equal to the preset image difference coefficient, it is determined that the photovoltaic power generation system to be monitored for fault has a fault risk.

8. A photovoltaic power generation system fault monitoring system, characterized in that: include: The first module is used to determine the photovoltaic power generation system to be monitored for faults, photograph the photovoltaic power generation system to be monitored for faults from multiple angles to obtain multiple multi-angle images, and stitch the multi-angle images to obtain a panoramic image to be analyzed for faults; The second module is used to perform grayscale conversion on the panoramic image to be analyzed for faults, obtain the panoramic grayscale image to be analyzed for faults, and determine a preset standard grayscale image of the photovoltaic power generation system to be monitored for faults; The third module is used to analyze the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image, and determine the image difference coefficient between the panoramic grayscale image to be analyzed for fault and the preset standard grayscale image based on the analysis result; The fourth module is used to determine whether the photovoltaic power generation system to be monitored for failure has a failure risk according to the image difference coefficient, and to issue an alarm reminder when the photovoltaic power generation system to be monitored for failure has a failure risk.

Images